Association of p-phenylenediamine exposure with alterations of pulmonary function, pruritus and health-related quality of life in hair dye factory workers: a cross-sectional study

P-Phenylenediamine (PPD) is a common component of hair dye, which can cause skin contact allergy and asthma with impaired pulmonary function. However, the adverse effects of occupational exposure to different dose PPD was rarely mentioned. We recruited 124 workers from a hair dye factory to explore the association of occupational PPD exposure on pulmonary function, pruritus and health related quality of life (HRQL). We categorized exposure to PPD into 3 levels: lower exposure group (< 0.00001 mg/m3); middle exposure group (0.00001–0.00033 mg/m3); higher exposure group (0.00033–0.047 mg/m3). The HRQL and subjective pruritus of the workers were assessed by the short form 36 health survey (SF-36) and Visual analogue scale (VAS) of pruritus, respectively. In the high PPD-exposed group, the percentage of FEV1 (FEV1%) was lower in higher exposure group compared with lower exposure group. The FEV1/FVC was also lower in comparison to the higher exposure and middle exposure groups (p < 0.05). PPD levels were negatively correlated with vitality and mental health (p < 0.01). The structural equation model showed the positive effects of PPD on VAS level (β = 0.213, p < 0.001), and indicated partly negtive effects of PPD on total score of SF-36 (β = − 0.465, p = 0.002), respectively. Our results indicate that occupational exposure to PPD might be associated with pulmonary function impairment, poor HRQL, and subjective pruritus of the workers.

PPD, is about 4.80% in outpatients with occupational exposure to PPD such as hairdressers, barbers 8 . PPD may also cause damage to the lungs. Induction routines and challenge patches with 0.001% to 10% PPD sensitized 56% to 100% of guinea pigs on test. Massive peribronchial infiltrates of eosinophils were observed in guinea pigs after intrapulmonary injection of an aqueous solution containing 1% PPD 9 . The serious level of PPD allergy is mostly determined by the exposure concentration and duration 10 . It is reported that PPD causes erythema, scaling, itching or edema in less than 1% individuals of the general population after dying the hair 5 . Moreover, long-term occupational exposure to PPD may lead to chronic renal impairment in workers 11 . Oxidized products of PPD might also have potential carcinogenicity to humans. Association has been observed between hair dye use and non-Hodgkin's lymphoma, multiple myeloma, acute leukemia and bladder cancer [12][13][14] .
As aforementioned, occupational exposure to PPD can result in health problems such as skin allergies and pulmonary function changes 6 . However, previous studies have mostly conducted research on ordinary people or emergency patients, and the studies of occupational groups only focused on individual case reports of hairdressers. So far, no research has been conducted on workers who are working with exposure PPD. And examination including peripheral blood, pulmonary function has been presented as an easily-used method reflecting the physical condition 15,16 . In addition, there are few studies evaluating the effect of the PPD exposure on the itching of the skin of the workers. In order to evaluate skin discomfort and correlate it with the effects on the individual, there is also a need for methods to determine the clinical severity of people. Compared with patch-test to PPD allergy in previous research, the Visual analogue scale (VAS) could evaluate the skin discomfort without an obvious allergic reaction 8,17 . Whether these adverse effects (pulmonary dysfunction and pruritus) of PPD lead to a decline in their life quality, and then interfere with their activities of daily living and work is noteworthy. Healthrelated quality of life (HRQL) is a broad and multidimensional measure that assesses an individual's perception of physiological function, mental ability and social adaptability. HRQL can be influenced by health status, medical interventions, personal beliefs, social environments, economic circumstances and cultural backgrounds 18 . The purpose of this study aimed to investigate the exposure status and evaluate pulmonary function of low dose PPD-exposed workers in hair dye production, and evaluate subjective pruritus and HRQL by the short form 36 health survey (SF-36) and Visual analogue scale (VAS) of pruritus, respectively. We believe that this study will provide useful information for health care policy makers and health care providers so as to develop better strategies on health education programs for PPD-exposed control.

Methods
Study population. The cross-sectional study was based on data from a hair dye factory containing PPD in Zhejiang Province of China. The occupational health management in this company is standardized, possessing protective measures that can prevent contact with dangerous substance includes masks, protective clothing, boots and overshoes. The factory employees were first coded and then randomly sampled by computer generated random numbers. The employees are engaged in configuring raw material (n = 9), adding chemical (n = 9), adding material into the jar (n = 28), packaging and transporting (n = 50), laboratory staffs (n = 4), accounting (n = 6), cleaning and logistics (n = 13) and managing (n = 5). The sample had 68 men and 56 women. No participants had history of allergic reaction. They were asked to fill out a demographic questionnaire which included the age, gender, area of residence, telephone number, working age, educational condition, smoking status and hair dye use status. All participants have signed informed consent, and ethical approval was granted by the Ethics Committee of Tianjin Center for Disease Control and Prevention.

Measurement of PPD concentration.
Referenced to the method 5013 of the U.S. National Institute for Occupational and Safety (NOISH) 19 , the PPD in the air of various working environments and workers of different work types (simple random sampling of 3-5 individuals per type of work) were sampled by Gilair-3 air sampling pump (Gilian, SENSIDYNE, St. Petersburg, FL, USA), collected with glass fiber filters coated with dilute sulfuric acid (Nantong Jinnan Glass Instrument, Jiangsu, China), and extracted with an aqueous EDTA solution (Guangfu Fine Chemical Research Institute, Tianjin, China), and then determined by high performance liquid chromatography (HPLC, SHIMADZU, Inc., Kyoto, Japan) with the analytical column of Eclipse SB-C18 (150 mm × 4.6 mm × 5 μm) and the mobile phase of mixed solution of disodium hydrogen phosphate solution and acetonitrile with a flow rate of 2.0 mL/min. The detection wavelength was 240 nm and the injection volume was 5 μL.
Self-reported measures of life condition. The subjects completed two self-reported assessment for life condition: SF-36 and VAS of pruritus.
The quality of life was assessed by SF-36, derived from the health survey results of the Medical Outcomes Study (MOS) in 1988. It has been translated into Chinese in 1991 and its reliability and validity has been verified in Chinese population after considering cultural adaptation, validation, and normalization 20 . SF-36 measured eight dimensions of health including physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional and mental health. In addition to these dimensions, SF-36 also considered reported health transition scaled scores 21 . Each scale was directly transformed into a score from 0 to 100 after modifying the weighing coefficient in each question. The lower score meant the worse health status, whereas the higher the score the less possible disability 22 . The Cronbach's α values for each dimension of SF-36 were calculated to be greater than 0.7, indicating that the internal consistency of the scale is good in occupational workers. After factor analysis, the KMO value was 0.625 and Bartlett's test result show p value was less than 0.001, indicating that the structural validity was good. Therefore, this scale had suitable reliability and validity for the occupational population. www.nature.com/scientificreports/ The extent of subjective pruritus was assessed by VAS of pruritus 23 . It has been widely used because of a simple structure and flexible operation. Participants could mark a score for a 100 mm horizontal line on the questionnaire to represent their subjective feelings of cutaneous itching. The benchmark descriptors for each level were as follows: 1 level: 0 mm to 10 mm (i.e., no itching symptoms), "no or almost no itch"; 2 level: 10 mm to 40 mm (i.e., slight itching symptoms), "occasional restless sensations that do not necessarily induce scratching behavior"; 3 level: 40 mm to 70 mm (i.e., moderate itching symptoms), "significant scratching and the itch disrupts sleep, and although the patient can go back into sleep after scratching"; 4 level: more than 70 mm (i.e., fierce itching symptoms), "unbearable itch causes scratching and scratching almost deprived of sleep, the patient keeps scratching ceaselessly". Because of the subjectivity of pruritus, we regard all patients with pruritus symptoms more than 0 mm as having pruritus symptoms.
Pulmonary function examination. The portable spirometer (Contec SP10, Contec Ltd., China) was used to measure the capacity and the volume of lungs..Measurement indexes included forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and ratio of FEV1 to FVC (FEV1/FVC), etc. Abnormal pulmonary function was defined as: FVC% (percentage of forced vital capacity, FVC measured value/FVC predicted value) < 80%; FEV1% (percentage of forced expiratory volume in 1 s, FEV1 measured value/FEV1 predicted value) < 75%; FEV1/FVC%, FEV1 measured value/FVC measured value < 70%. The predictive values of FVC or FEV1 was automatically calculated by the pulmonary function analyzer according to age, gender, height, weight and race. A disposable mouthpiece was used for each participant. The participant inhaled maximally through the nose until the lungs were full. Afterward, the participant placed the spirometer through the disposable mouthpiece in his/her mouth, with lips sealed tightly around the mouthpiece while holding the lungs full 24 . The participant was instructed to exhale forcefully as long as possible into the spirometer until no air could be exhaled. A minimum of three trials was done and the values of FEV1, FVC and FEV1/FVC were obtained. Statistical analysis. Statistical analysis was conducted using SPSS package (Version 20.0) unless otherwise specified. Numerical variables were reported as mean ± SD. The Chi-square test and Chi-square after Yates continuity correction test were used to compare the differences of demographic situation and working age, educational, smoking and hair dye use between different groups. The covariance analysis and Bonferroni test were performed for comparing interesting variables in different groups after controlling for the confounding factors, and adjusted P < 0.017 (two-tailed) was considered statistically significant, taking the multiple comparisons into consideration. Covariates included gender, ethnic group, working age, education condition and recent hair dye history. Due to the co-linearity of working age and age, this study only adjusted working age as a covariate. The influencing factors of VAS levels in different groups were studied by logistic regression. Before the multivariate analysis, the Variance Inflation Factor was calculated to rule out the effects of multicollinearity. We performed correction test of collinearity and used logistic regression to estimate the odds ratio (OR) and its 95% confidence interval (95% CI) for the association of VAS level with hematologic indexes and some factor scores in SF-36. p < 0.05 (two-tailed) was considered statistically significant in the final models. Structural equation modeling (SEM), which can evaluate the relationship between variables, was performed with the AMOSE software. Cutoff criteria for good model fit were attained from extant literature; Satorra-Bentler χ 2 /df values < 3.0 were considered good; Goodness of Fit Index (GFI) values > 0.90 and Adjusted Goodness of Fit Index (AGFI) > 0.90 were considered reasonable; Root Mean Square Error of Approximation (RMSEA) values < 0.10, Incremental Fit Index (IFI) and Comparative Fit Index (CFI) values > 0.90, parsimony goodness-of-fit index (PGFI) > 0.50 were considered good. Consistent Akaike Information Criterion (CAIC) default < CAIC independence and CAIC default < CAIC saturated were considered good 25 . Missing data was replaced by means of the scores in the different groups.
Approval of the research protocol. All procedures performed in the study involving human participants were in accordance with the ethical standards of the Ethics Committee of Tianjin Center for Disease Control and Prevention and with the 1975 Declaration of Helsinki and its later amendments or comparable ethical standards. The study was approved by the Ethics Committee of Tianjin Center for Disease Control and Prevention.
Informed consent. Informed consent was obtained from all individual participants included in the study or their legal guardians.

Results
General characteristics of study population. The  www.nature.com/scientificreports/ workers who were involved in raw material, adding chemical, adding material into the jar, and laboratory staffs were designated as the higher exposure group (0.00089-0.19 mg/m 3 ) (Supplemental Table 1). The distributions of demographic characteristics among the three groups were showed in Table 1. Higher exposure group, middle exposure group and lower exposure group statistically differed in gender, ethnic group, working age, education condition and hair dye history (P < 0.05). The male constituent ratio was higher in the higher exposure group (82%) than in the middle exposure group (36%) and the lower exposure group (37.5%). Among the three groups, the proportion of the Han nationality in middle exposure group (80%) was lower than the other groups. The number of workers with working age less than 5 years was the highest in the middle exposure group (54%). In the term of educational condition, the proportion of primary school and lower level was lower in the lower exposure group than the other two groups. The hair dye history in the lower exposure group (16.67%) had a lower constituent ratio than the higher exposure group (54%).
Changes of hematologic indexes of the study participants. The hematologic indexes results among the three groups were in the normal count ranges. After adjustments for gender, ethnic group, working age, education condition and recent hair dye history, the covariance analysis showed that in the higher exposure group, total white blood cell count was lower (p < 0.001), while basophils and eosinophils were higher (p < 0.05) than that in the lower and middle exposure groups. Besides, compared to the middle exposure group, workers in the higher exposure group presented a higher neutrophil and hemoglobin level. No differences were observed for other hematologic indexes indices among the three groups ( Table 2).
Alteration of pulmonary function of the study participants. After adjustments for gender, ethnic group, working age, education condition and recent hair dye history, the spirometry parameters of the three groups were significantly different. The pulmonary function indices like FEV1% and FEV1/FVC% of the higher exposure group were significantly lower than those of the lower exposure group (p < 0.05). In the higher exposure group, FVC% was higher, while FEV1/FVC% was lower than that in the middle exposure group (p < 0.05). In addition, compared to the lower exposure group, workers in the middle exposure group presented the lower FVC% and FEV1% (p < 0.05) ( Table 3). www.nature.com/scientificreports/ Self-reported of physical subjective feelings of the study participants. After excluding the influence of gender, ethnic group, working age, education condition and recent hair dye history, covariance analysis indicated that means for role-physical (p = 0.023), bodily pain (p < 0.001), general health (p < 0.001), vitality (p < 0.001), mental health (p < 0.001), health transition (p < 0.001) and total scores were all significantly different among the three groups (Table 4). The means for bodily pain, general health and health transition in the middle exposure group were lower than the lower exposure group and the higher exposure group (p < 0.05). Besides, the mean for mental health in the higher exposure group and role-physical in the middle exposure group was lowest of all groups (p < 0.05). Higher concentration PPD also resulted in a decreased vitality score compared to the lower exposure group. For the total score, the score for the lower exposure group was higher than the other groups, and the higher exposure group was higher than the middle exposure group (p < 0.05).
The results of multi-variable logistic regression analysis for VAS levels with some demographic variables, SF-36 factors were showed in Table 5. It found that men were less likely to present high VAS levels than women (OR 0.114, 95%CI 0.027, 0.482, p = 0.003). The worker who works in a higher concentration PPD atmosphere significantly increased the probability of VAS level (OR 9.394, 95%CI 1.710, 51.622, p = 0.010). Individuals without a hair dye history had almost one tenth the probability of not (OR 0.091, 95% CI 0.021, 0.385, p = 0.001). And workers with senior middle school and higher education have lower level of VAS (OR 0.041, 0.033 respectively; both p < 0.05). As for SF-36 scores, workers with the higher BP score have higher VAS level (OR 1.056, 95% CI 1.008, 1.106, p = 0.021), and with higher GH score have lower VAS level (OR 0.950, 95% CI 0.904, 0.998, p = 0.040). Table 2. Covariance analysis results of serum biochemical indexes (mean ± SD). The covariant variables including gender, ethnic group, working age, education condition and recent hair dye history. *P < 0.05, comparison of the higher and middle exposure group with the lower exposure group by Bonferroni method. # P < 0.05, comparison between the higher exposure group and middle exposure group by Bonferroni method.

Discussion
In our study FEV1% was lower in higher exposure group compared with lower exposure group. The FEV1/FVC% was also lower in comparison to the higher exposure and middle exposure groups. PPD levels were negatively correlated with vitality and mental health. The structural equation model showed the positive effects of PPD on VAS level, and indicated partly negtive effects of PPD on total score of SF-36, respectively. Occupational PPD exposure might be associated with pulmonary injury, poor HRQL, and subjective pruritus of the workers. Frequent contact with hair dyes production and working in the atmosphere with PPD aerosol can caused widespread health issues such as contact allergy 26 . In the present study, the blood biochemical indices, pulmonary function, HRQL and VAS of pruritus were compared among the workers who exposed to low dose PPD aerosol, because the results of air monitoring indicated that the mean concentration of PPD in different workshops were lower than the threshold-limited value-time-weighted average (0.1 mg/m 3 ). The workers exposed in higher PPD concentration showed a lower white blood cell count and higher basophils, hemoglobin and eosinophils counts. An elevation of basophils, hemoglobin and eosinophils counts was similar to those found in neutrophilic dermatitis, apart from the pronounced eosinophilia. Numerous studies have reported that is a marker of hypersensitivity [27][28][29] . Moreover, the results of the present work agreed with most findings of the Sieben's study that both PPD and its auto-oxidation product may stimulate lymphocyte proliferation, which is an important marker of inflammatory responses as well as allergic reactions and confirms the allergenicity of PPD 30,31 .
Our study was the first to explore whether PPD exposure induced the changes of pulmonary function examination. However, the sample size of our study was relatively too small, and the conclusion remains to be further verified. Both FEV1% and FEV1/FVC% were lower in the higher PPD exposure group, and FVC% was lower in the middle PPD exposure group than other two groups which are all more than 70% and are at normal levels 32 . In addition, the mean of all groups' indices were in the normal range. The reason may cause this situation was that the hazard of low PPD dose with long-term exposure was relatively weaker. Some case report researches have also indicated that PPD poisoning could lead to respiratory distress 33,34 . However, these studies have not reported the effects of low PPD concentration on pulmonary function in the workplace. This finding implied that there might be an increased risk of respiratory disorders for workers in response to long-term PPD exposure.
Any health problems can cause a disruption in the normal life and have a substantial impact on the quality of life 35 , it seems important to assess the non-organ-specific manifestations such as HRQL. In the present study, we found that not only body pain, mental health and physical role were affected, but also the general health, vitality and reported health transition were affected for all workers. It has been suggested that these effects can have a long term effect of the acquisition of personality and health behaviors. Our results were similar to the conclusions reached by Mohammadi et al., who found that the workers exposed to higher levels of crystalline silica had lower values of pulmonary function indices and lower health-related quality of life 36 . Various studies have showed that age is a risk factor to physical health status 37,38 , because elders were usually affected by one or more chronic diseases and conditions 39 . However, the present study showed that the general health status, vitality and health transition were better in longer work-age workers. It might be explained by the healthy worker effect phenomenon that older workers usually have better health conditions than the general population 40,41 . People exposed in higher PPD levels have lower score in the dimensions of vitality and mental health. One reason for the results might because they engaged in different types of work that caused different PPD exposed concentration. The other reason for this could be the different ways to select subjects, with studies in this field based on retrospective analysis. The current study found that subjective feelings of bodily pain, mental health, and general health are not the same for men and women. As reported by previous study, general health and mental health were significantly associated with gender 10 .
In the light of the itching of skin seen as a subjective discomfort, the VAS of pruritus was used in this survey. The results of VAS levels in this survey are related to PPD exposed levels. A similar tendency was found that PPD is responsible for the majority of the allergic reactions to permanent hair color, among which the most common kind of allergic reactions is itchy 42 . It found that women were more likely to feel itching of skin than men. This is partly due to the physiological sensitivity of different gender to PPD after control the influence of gender 43 . Because VAS level is also affected by subjective perception, we analyze the impact of health-related quality of life and education condition on VAS level. It found that the higher the level of education, the lower the degree of itching. This may because that a high level of education will have a more comprehensive understanding of the dangers of PPD. The reciprocal and intricate relationship between the psyche and itch has been widely studied 44 . The augmentation of perception of itch owing to the interplay of psychogenic and emotional factors has been well studied 45 . Although pruritus itself is not recognized as a separate entity as such, DSM-V (Diagnostic and Statistical Manual of Mental Disorders, 5th edition by American Psychiatric Association) has already included a category of physical symptoms and related disorder. Pruritic conditions exacerbate the symptoms of itching. Itching and pain have the same receptors and have much in common 46 . Therefore, there is a certain relationship between the degree of body pain and VAS level.
We found direct paths from PPD exposure level to VAS level outcomes through one variable: influence of SF-36. Specifically, have hair dye history predicted negative affect on VAS level. PPD exposed may cause FEV1% and FEV1/FVC% decreased, and some influence to blood biochemical index such as white blood cell count, eosinophils, basophils and hemoglobin according to the results of structural equation model.
A key strength of our study is that we first studied the workers with the occupational exposure of low dose PPD as the research object, and measured the levels of occupational exposure to PPD in groups. Therefore, the www.nature.com/scientificreports/ exposure grouping in this study is more accurate and reasonable. An additional strength of our study is we comprehensively consider the physical and mental health of workers. This study analyzed the results of pulmonary function and HRQL according to the level of exposure. The results of our study also provided some useful data to support the formulation of occupational exposure limits (OELs) for PPD in workplace air, which has been included and issued as "Occupational exposure limits for hazardous agents in the workplace. Part 1: Chemical hazardous agents GBZ2.1-2019". Further effective guidelines and measures will certainly be provided for the health protect of occupational population in China 47 .
There are some limitations in this study. First, the study is a cross-sectional study. Although a possible causal association has been established through structural equation modeling, further cohort studies and experimental studies are needed to verify these association. Secondly, the study was limited by its small sample size. Thus, large-scale studies are necessary in PPD occupationally exposed workers to gain a better understanding of the physiological and mental state of the workers. Moreover, there were different types of confounding factors for SF-36 scores and VAS levels: the socio-economic status, income, work type, and stress for life, therefore, further comprehensive studies should consider these factors.

Conclusion
Our current study showed that occupational PPD exposure has an effect on subjective itching and health-related quality of life, and health-related quality of life mediates the effects of PPD exposure on subjective pruritus.

Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.